Repository navigation
neuron_segneuron: compare with SegNeuron in the paper's VI convention - #229
Merged
Merged
Conversation
…age 0.17.2) The paper's Table 3 VI divides the GT-0-ignored contingency table by the total voxel count (yanchaoz/SegNeuron#8), so it is the current skimage VI times the GT foreground fraction. evaluate_emneuron.py writes vi_old and gt_fg_frac; the tutorial table compares in that convention. Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The SegNeuron authors clarified (yanchaoz/SegNeuron#8) that Table 3 was scored with scikit-image 0.17.2, which ignores GT 0 but normalizes by the total voxel count: VI_old = VI x count(GT != 0) / GT.size; ARE is unchanged.
scripts/evaluate_emneuron.py: two new TSV columns,vi_oldandgt_fg_frac(existing columns unchanged).tutorials/neuron_segneuron/README.md: the results table compares in the paper's convention and keeps the current-skimage VI beside it.Checks:
vi_oldequals an independent computation with the 0.17.2 normalization on a random volume; with the measured foreground fractions the released SegNeuron checkpoint + FRMC reproduces Table 3 (0.8655, 0.7718, 1.0230 at beta 0.30 on Harris vs 0.8655, 0.7719, 1.0221). Table values are the existing scores times the foreground fraction; nothing was re-decoded.🤖 Generated with Claude Code